Adam Day is a Data Scientist at SAGE Publishing. He uses Natural Language Processing and graph analytics to detect research misconduct.
Ebru Cucen is a lead consultant at OpenCredo. She is a hands-on architect, and she has been involved in the end-to-end design, development and delivery of scalable fault tolerant cloud-native solutions. With a more recent focus on data engineering, graphs and ML/AI, she looks to combine the best of both worlds - gaining new and fresh insights, delivered through a stable, reliable solution. She is the author of the free ebook Graph ML book explaining use cases for Graph. [https://opencredo.com/connect-the-dots-harness-the-power-of-graphs-ml-ebook/ ]
Searching for research fraud in OpenAlex with Graph Data Science
The OpenAlex dataset is a vast graph of data describing researchers, research papers, institutions, journals and more. It is known that some organisations exist which offer authorship slots on research papers for sale. Authors of research papers are required to contribute to the research described in the paper, so paying for an authorship slot is a form of research misconduct. Given a dataset of research papers, where it is known that authorship slots were sold in this way, we use OpenAlex and Neo4J hosted on Google Cloud Platform to analyse data relating to these papers and take a first step towards automated detection of such unusual co-authorships.